研究目的
To develop an automated system for analyzing digital fundus images and diagnosing diabetic retinopathy for early detection and prevention of blindness.
研究成果
The developed system can be used by biomedical engineers to help doctors and remotely resident patients. If developed to work online, it can identify eye conditions of patients far from hospitals, potentially serving as a kiosk in remote areas for maximum coverage.
研究不足
The system is currently a prototype in software model only; development in a hardware model requires sufficient funding.
1:Experimental Design and Method Selection:
The study involves the development of a Graphical User Interface (GUI) for the detection of diabetic retinopathy using image processing techniques.
2:Sample Selection and Data Sources:
Retinal images of patients were taken from Regional Institute of Ophthalmology, Medical College, Thiruvananthapuram.
3:List of Experimental Equipment and Materials:
Fundus Camera for obtaining fundus images.
4:Experimental Procedures and Operational Workflow:
Image enhancement using filtering, segmentation to detect optic disc, fovea, exudates area and blood vessels, and use of connected component method along with concentric circle methods to determine the artery-vein width ratio.
5:Data Analysis Methods:
Development of an algorithm for the detection and quantification of the disease level from the parameters specified, validated with clinical data.
独家科研数据包,助您复现前沿成果,加速创新突破
获取完整内容